表題番号：2020C-605
日付：2021/02/11
研究課題速報的情報を俯瞰するためのテキストの集約と分析手法に関する研究
| 研究者所属（当時） | 資格 | 氏名 | |
|---|---|---|---|
| （代表者） | 理工学術院 基幹理工学部 | 教授 | 河原 大輔 |
- 研究成果概要
- As the coronavirus disease (COVID-19) has been rapidly spreading around the world, there is an increasing need for a system for aggregating immediate information that transcends borders and domains. To build such a system, it is necessary to use natural language processing (NLP) technologies flexibly, such as combining machine translation of multilingual texts with information analysis technology, and mapping information transmitted by experts with social media texts. We have studied an application of NLP technologies for COVID-19 by cooperating with researchers in informatics including NLP. Then, we have developed a system for aggregating COVID-19 information from all over the world. In this system, COVID-19 information is grouped by regions and topics, such as infection status, prevention, medical information, economic policies, and education. Collected multilingual articles are translated into Japanese and English by machine translation and are automatically classified into the topics by the contextualized language model BERT. We hope that this system is useful for many people, and this kind of technology will be used for other future events and disasters.